Abstract
Background/Aim: Tubular adenomas of the colon (TA) are neoplastic polyps composed of dysplastic tube-like crypts. Nondysplastic crypts, mostly in asymmetric branching have been previously reported, both beneath and bordering TA. In the present article, intercalated nondysplastic crypts (INDC) amidst dysplastic crypts in TA are showcased. Patients and Methods: The occurrence of INDC was recorded in 139 TA. Results: Out of the 139 TA, 31% exhibited INDC; of these, 58% were in asymmetric branching (INDCAB), 35% were single intercalated crypts without branching (INDSNB), and 7% were in symmetric branching (INDCSB). Luminal dysplasia occurred in 53% out of the 43 TA: in 37% TA with INDCAB, in 16% TA with INDSNB, but in none of the TA with INDCSB. Thus, INDCAB predominated. Conclusion: The finding of INDC in TA domain contrasts with the infrequency of INDCSB and with the absence of INDCAB in the normal colorectal mucosa. Hence, INDC emerge as integral components in TA. Since only 1 or 2 sections were available per TA, the total number of INDC in the entire TA is likely higher. INDC in TA may be remnants of acquired nondysplastic mucosal cores of abnormal cryptogenesis that were subsequently replaced by top-down growing dysplastic epithelium. The present and previous findings support the concept of field cancerization in the human colorectum.
Based on the observation that many oral neoplastic lesions recurred after apparent radical extirpation, Slaughter et al. launched the concept of field cancerization 71 years ago (1). Slaughter’s conclusions included microscopic multicentric origin (as deduced by serial sections) in tumours less than 1 cm in diameter, and abnormal and hyperplastic, often atypical, epithelium surrounding oral cancers (1). Currently, the field cancerization concept asserts that the normal tissue adjacent to the tumour harbour certain preneoplastic genetic fingerprints, which can eventually lead to the development of local recurrence or secondary primary tumours (2). Field cancerization can occur in the colon (2).
In previous morphologic research on field cancerization in the colon, we reported the presence of nondysplastic crypts in asymmetric branching underneath and neighbouring polypoid adenomas (3, 4), and nonpolypoid adenomas (5, 6). It should also mentioned that six out of 80 nonpolypoid adenomas investigated revealed nondysplastic branching crypts interspersed amidst the dysplastic glands (4). It was hypothesized that the various localizations of nondysplastic branching crypts in nonpolypoid adenomas represented remnants of an acquired mucosal hub having abnormal cryptogenesis.
The purpose of this communication was to explore whether nondysplastic branching crypts could also occur interspersed among the dysplastic glands of polypoid tubular adenomas (TA) of the colon, the most prevalent of all colon adenomas.
Patients and Methods
Patients. The material includes 500 colon adenomas retrieved from the electronic archive (DC Pathos 8 database) of the Institute of Pathology, Friedrich-Alexander Universität Erlangen- Klinikum Bayreuth, Germany. The sections (4 μm) stained with Hematoxylin-eosin (H&E) were digitalized and scanned using a Hamamatsu NanoZoomer Digital Pathology S360 (NDP, Hamamatsu, Herrsching am Ammersee, Germany). A web interface made images accessible to all authors. Out of the 500 colon adenomas, 305 were TA. Since an important prerequisite for inclusion was that sections were unfragmented, only 139 TA out of the 305 TA were considered suitable for the study.
Assessing the size of tubular adenomas. Measurements were performed on images of TA generated by the Hamamatsu NanoZoomer Digital Pathology S360 (NDP, Hamamatsu, Herrsching am Ammersee, Germany). The broadest TA diameter was logged using the digital ruler adjusted in mm. When ≥2 histologic phenotypes of intercalate crypts/TA were found, TA exhibiting intercalated crypts in asymmetric branching were registered as such.
Definitions. Intercalated crypts: Test tube-like nondysplastic crypt having the open luminal orifice sandwiched between the dysplastic crypts of TA.
Intercalated single crypts: Test-tube-like nondysplastic crypt without branching with the open luminal orifice interposed among dysplastic crypts of TA (Figure 1).
Intercalated nondysplastic single crypts without branching amidst dysplastic crypts of colon tubular adenomas. Upper panel. Intercalated nondysplastic single crypts without branching. A, B, and C: H&E, original magnification: A and B ×10, C ×20. Lower panel. Intercalated nondysplastic single crypts without branching exhibiting dysplasia in the upper luminal aspect of the crypts D, E, F. H&E, original magnification: D, E and F ×10.
Intercalated crypt in symmetric branching: Nondysplastic crypts interposed amidst dysplastic crypts of TA showing a luminal orifice opening, a pre-branching sector and two post-branching identical offsprings (Figure 2A).
Intercalated nondysplastic crypts in asymmetric branching amidst dysplastic crypts in colon tubular adenomas. Upper panel. Intercalated nondysplastic crypts in asymmetric branching. H&E, original magnification: A and B ×10, C ×4. Lower panel. D, E, F. Intercalated nondysplastic crypts in asymmetric branching exhibiting dysplasia in the upper luminal aspect of the crypts (H&E, original magnification: D ×4, E ×10 and F ×10).
Intercalated crypt in asymmetric branching: Nondysplastic crypts interposed amidst dysplastic crypts of TA showing a luminal orifice opening, a pre-branching sector and two or more post-branching offsprings of different shapes (Figure 2B and C).
Luminal dysplasia in intercalated crypts: Epithelial dysplasia present in the upper portion of intercalated crypts (Figure 1D, E, and F, Figure 2D, E, and F, and Figure 3).
Intercalated nondysplastic crypts amidst dysplastic crypts in colon tubular adenomas. Upper panel: A and B Tubular adenomas exhibiting intercalated nondysplastic crypts amidst dysplastic crypts at arrows (H& E, ×1). Lower panel: C and D, Higher magnifications, corresponding to the arrows in A and B, respectively (H&E, original ×20).
Ethical approval. Ethical approval was obtained from the Ethics Committee of Friedrich-Alexander University, Erlangen-Nuremberg, Germany, for adenomas (ID number: 21-475-Br).
Statistical analysis. The nonparametric Mann-Whitney U two-tailed test was applied to compare the difference between two groups. Statistical significance was defined as p<0.05.
Results
Out of the 139 TA, 43 TA (31%) thrived with intercalated crypts without or with dysplasia on top amidst the dysplastic crypts of the adenomas. Table I shows that 58% of the TA had intercalated crypts in asymmetric branching, 35% intercalated single crypts without branching, and the remaining 7% had intercalated crypts in symmetric branching. Thus, intercalated crypts in asymmetric branching predominated. The Table also shows that luminal dysplasia was present in 53% (n=23) of the 43 TA: in 37% of the TA with intercalated crypts in asymmetric branching and in 16% of the TA with single crypts without branching, and in none of the TA with intercalated crypts in symmetric branching. Thus, luminal dysplasia in intercalated crypts in asymmetric branching, predominated.
Frequency of intercalated crypts in 43 tubular adenomas (TA). Percent in brackets.
Since some TA had ≥2 intercalated crypts/adenoma. the frequency of all intercalated crypts appearing in the 43 adenomas was recorded. Table II shows that out of the 50 intercalated crypts, 54% were in asymmetric branching, 36% were single crypts without branching and the remaining 10% in symmetric branching. The Table also shows that dysplasia at the luminal aspect was present in 50% of all intercalated crypts; in 63% of those in asymmetric branching, in 44% of the intercalated single crypts without branching, but in none of five intercalated crypts in symmetric branching. Hence, the predominant crypt phenotype of intercalated crypts with luminal dysplasia exhibited asymmetric branching.
The frequency of nondysplastic intercalated single crypts without branching and of intercalated crypts in asymmetric and in symmetric branching in 43 tubular adenomas (TA) of the colon. The frequency of nondysplastic intercalated crypts with dysplasia on top is also shown (percent in brackets).
The mean size of the 43 TA having intercalated crypts was 5.70 mm is shown in Table III. The mean size for the 25 TA with intercalated crypts in asymmetric branching was 6.45 mm, for the 15 TA with single intercalated crypts without branching 4.27 mm, and for three TA with intercalated crypts in symmetric branching 6.33 mm. The few numbers of TA with intercalated crypts in symmetric branching disallowed statistical challenge. The difference between the first two groups was significant at p<0.05 (p=0.0027; Mann-Whitney U two-tailed test). Thus, the size of TA having intercalated crypts in asymmetric branching was significantly larger than the size of TA carrying single intercalated crypts without branching. The size in the 23 TA with intercalated crypts exhibiting luminal dysplasia is also summarized in Table III. The mean size in the 16 TA with luminal dysplasia in intercalated crypts in asymmetric branching was 6.96 mm, and in the seven TA with luminal dysplasia in single intercalated crypts without branching 5.66 mm. The difference in size between those two groups was significant at p<0.05, (p=0.0027; Mann-Whitney U two-tailed test). Although the size in the three TA with intercalated crypts in symmetric branching having luminal dysplasia is shown in Table I, the few numbers of TA in that subgroup disallowed statistical evaluation.
The size in mm (range in brackets) in 43 tubular adenomas (TA). The size in intercalated single crypts without branching, in intercalated crypts in asymmetric branching and in intercalated crypts in symmetric branching are shown.
Discussion
This survey showed that 31% of the TA investigated concurred with nondysplastic crypts intercalated amidst dysplastic crypts. Luminal dysplasia was found in 53% of the intercalated crypts. The most predominant histologic phenotype amongst intercalated crypts was that exhibiting asymmetric branching s (without or with luminal dysplasia). Importantly, TA exhibiting intercalated crypts in asymmetric branching s were larger in size. The possibility that the intercalated single crypts without branching found in smaller TA could subsequently proceed to a branching mode in larger TA cannot be completely disregarded. Considering that only one or two H&E-stained sections were available for the study, it appears safe to postulate that the frequency of intercalated crypts here reported may reflect only an aliquot of the total number of intercalated nondysplastic crypts that supposedly are present in the entire TA. This is remarkable considering that in the normal colon mucosa (7, 8) symmetric branching crypts are uncommon (9), and crypts in asymmetric branching do not occur. This assertion is also valid for the normal colon mucosa in rodents (10).
In larger series of colorectal adenomas, the frequency of solitary adenomas ranged from 66% out of 3,135 (11) and 67% out of 3,037 (12) to 75% out of 2,506 (13) adenomas. Apparently, colorectal adenomas often evolve as solitary polyps. In this context, it should be mentioned that driver mutations have been found in around 1% of normal colorectal crypts in middle aged individuals (vide infra). It is conceivable that isolated colon cores of nondysplastic branching crypts are the site from which dysplasia and eventually TA develop. In an earlier experiment (10), we attempted to answer that question. The study was carried out on 35 male rats: 25 treated with the colonic carcinogen 1,2-dimethyhydrazine (DMH), suspended in EDTA solution, and 10 EDTA-treated. H&E-stained Swiss-roll sections revealed thousands of normal colon crypts as well as hubs of nondysplastic colon crypts with corrupted shapes in fission (i.e., branching) having either asymmetric basal fission, asymmetric lateral fission, asymmetric lateral sprouting, crypts with basal dilatation (≥ than twice the diameter of the normal lumen), or crypts with spatial aberrations of the normal (vertical) axis. Evidently, some of those morphological characteristics harmonized with the current concept of crypts in asymmetric branching (14). In DMH-treated rats, crypts with corrupted shapes in fission were found in 38% of the 533 colon crypts in fission but only in 0.1% of the 571 colon crypts in fission in EDTA-treated rats. Thus, hubs of nondysplastic crypts with corrupted shapes in fission developed in the colonic mucosa of carcinogen-treated rats. Importantly, the H&E-stained Swiss-rolls revealed four TA. Dysplastic epithelium was found replacing those hubs of acquired nondysplastic crypts with corrupted shapes in fission, in a top-down manner. The finding that crypt branching with corrupted shapes preceded the early stages of colonic adenogenesis in carcinogen-treated rats (10) strongly substantiate the results presented in this survey for TA in humans. Trying to learn more about the significance of intercalated crypts in TA, basal questions regarding colon crypts should first be addressed:
i) Are intercalated nondysplastic branching crypts in TA in humans being replaced by dysplastic epithelium?
According to the current pathway of top-down replacement and extension in human colon adenomas launched by Shih et al. in 2000 (15), dysplasia in adenomas evolves at the luminal aspect of adenomas and eventually progresses downwards. The authors also found that the dysplastic cells at the summit of the crypts displayed neoplasia-associated gene expression patterns and genetic changes of adenomatous polyposis coli (APC). Shih et al. proposed a top-down mode of epithelial replacement by genetically modified cells migrating laterally and downward to replace the epithelium of preexisting normal crypts. The authors claimed that the mechanism of replacement lead to the subsequent development of adenomas (15).
ii) Do normal colon cells thrive with somatic mutations?
It is generally accepted that the entire colon mucosa is constantly exposed to a series of oncogenic factors. Solitary or multiple colorectal adenomas may be the result of a protracted exposure to harmful factors (16). In 2003, Yang et al. (17) found that aneuploidy in normal neural progenitor cells was caused by abnormalities in chromosomal segregation. Tomasetti, Vogelstein, and Parmigiani (18) found that more than half of the mutations in certain malignancies were also detected in the normal cells from which the cancer had developed, concluding that those somatic mutations would have existed even in the absence of the tumour (18). According to Lee-Six et al. (19) and Dou et al. (20) normal cells from a variety of tissues can accumulate hundreds to a few thousand substitutions without becoming hypermutated. Up to 10% of these mutations may be detrimental, but most of them have no negative effects (20). Based on those findings, it appears fair to postulate that nondysplastic crypts in TA may harbour somatic mutations. iii) What causes nondysplastic colon crypts to alter their pristine tube-like shape?
In recent years, much research has been focused on the role of multifunctional growth factors (known as bone morphogenetic proteins, BMPs) in embryonic development and cellular functioning in postnatal and adult animals. Itoh et al. (21) demonstrated in three-dimensional cocultures in vitro that BMP2 (bone morphogenetic protein 2, similar in structure to TGF β) stimulated mouse stomach epithelium to form branching duct-like structures. This finding raises the possibility that BMP2 morphogen protein could be a morphogenic candidate for generating crypts in asymmetric branching in the colon mucosa.
Conclusion
Intercalated nondysplastic crypts were found in nearly one third of the TA and those in asymmetric branching were the predominant phenotype. The finding of intercalated nondysplastic crypts in symmetric and in asymmetric branching in TA domains contrasts with the infrequency of symmetric branching crypts and with the absence of asymmetric branching crypts in the normal colorectal mucosa. Intercalated nondysplastic crypts emerge as integral components in some TA. Since only 1 or 2 sections were available per TA the total number of INDC in the entire TA should be higher than those shown in Results. Intercalated nondysplastic crypts in TA may be remnants of acquired nondysplastic mucosal cores of abnormal cryptogenesis. Intercalated nondysplastic crypts were subsequently replaced by top-down growing dysplastic epithelium, The present findings support the concept of field cancerization in the human colorectum.
Footnotes
Authors’ Contributions
CAR is responsible for conceptualization, conducting the project, visualization, writing the original draft, and data curation, formal analysis, investigation and methodology, CL-S and MV for the scanning of sections. MV, CL-S, CM and KK for participating in the counting reliability quiz and for reviewing and editing the original draft. The final draft was approved by all Authors.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
Funding
This research did not receive any grants from funding agencies in the public, commercial, or not-for-profit sectors
- Received August 9, 2024.
- Revision received August 26, 2024.
- Accepted August 27, 2024.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.